Περίληψη :
Imaging in radio astronomy is a hard ill-posed inverse problem, particularly within the framework of very-long baseline interferometry (VLBI). The extremely sparse sampling of the uv-plane inherent to VLBI complicates image reconstruction and results in non-unique images. With the advent of cutting-edge telescopes such as the Event Horizon Telescope (EHT) and the next-generation EHT (ngEHT), the development of sophisticated imaging algorithms has become a critical area of research. Specifically, algorithms are being developed not only to produce static images but also to generate dynamic reconstructions—effectively, time-resolved movies—of astronomical sources, notably Sagittarius A* (SgrA*).
In this talk, I will provide a comprehensive overview of recent advancements in various imaging algorithms, discussing their underlying principles, strengths, and limitations, with a particular emphasis on those utilized by the EHT Collaboration to reconstruct images from challenging observational data. Following this, I will examine recent progress in regularized maximum likelihood (RML) methods to obtain full Stokes dynamic reconstructions (Müller & Mus+23, Mus & Müller+24a, Mus & Martí-Vidal+24b, Mus+24c) and that address critical limitations of earlier approaches by using multiobjective optimization and nature inspired optimization algorithms.
Additionally, I will present findings from a recently submitted study (Mus+Sub.), demonstrating how these algorithms can be employed to recover not only the astronomical sources but also the scattering screens that distort observations—an aspect particularly relevant for Sgr A* and Galactic Center observations. I will discuss novel perspectives on resolving the ring-like structure of Sgr A* at 86 GHz using these algorithms, a problem previously considered impossible.
